arcee-ai · First MLX build 2 Apr 2026
Trinity-Large-Thinking
MLX
Text apache-2.0
A 399B parameter text generation model from arcee-ai, with 2 MLX builds, from 4-bit to 5-bit.
- Parameters
- 399B From Hugging Face
- Quantisation
- 4-bit Trinity-Large-Thinking-4bit
- Memory needed
- 253 GB Estimate, unified memory
- Download
- 224 GB Weights, estimated
- Downloads
- 37 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 253 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Trinity-Large-Thinking-4bit --prompt "Hello"MLX builds
2 on mlx-community| Trinity-Large-Thinking-4bit | 4-bit | 224 GB | 253 GB | 37 |
|---|---|---|---|---|
| Trinity-Large-Thinking-5bit | 5-bit | 274 GB | 309 GB | 27 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.